Tau-Bench Airline — leaderboard
Metric: Pass@1 (%). Source: taubench.com. 16 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro | 73 |
| 2 | Claude Sonnet 4.5 | 70 |
| 3 | Qwen 3 Max (Thinking) | 69 |
| 4 | Claude 3.7 Sonnet | 64.2 |
| 5 | DeepSeek V3.2 | 63.8 |
| 6 | GPT-5 | 62.5 |
| 7 | Claude Sonnet 4 | 60 |
| 8 | Claude Opus 4 | 59.6 |
| 9 | Qwen 3 Max | 59.5 |
| 10 | Kimi K2 | 56.5 |
| 11 | GPT-4.1 | 56 |
| 12 | Claude Opus 4.1 | 56 |
| 13 | O4 Mini | 52.1 |
| 14 | O3 | 52 |
| 15 | GPT-4.1 Mini | 48.7 |
Interactive version: theaggregate.ai/benchmark?slug=tau-bench-airline · How the rankings work · Data refreshed daily, snapshot 2026-07-22.